protein sorting within the eukaryotic secretory 
pathway. 
Protein secretion is a fundamental and evolution- 
arily conserved process in eukaryotic cells. This pro- 
cess has been studied in baker's yeast by the labora- 
tory of Investigator Randy W. Schekman, Ph.D. 
(University of California, Berkeley ) by taking advan- 
tage of facile genetic and molecular cloning tech- 
niques together with the prospect of large-scale 
biochemical characterization. Genes and gene prod- 
ucts have been identified that are implicated in the 
early phases of protein secretion. New membrane 
proteins are believed to form a complex that allows 
secretory proteins, which generally are water solu- 
ble, to pass through a water-insoluble membrane. 
The proteins implicated in this process may be ex- 
tracted from membranes using mild detergents and 
then reconstituted into artificial membranes that re- 
produce the translocation process. A large and com- 
plex family of genes has been discovered that gov- 
erns the intracellular movement of proteins 
contained within membrane-enclosed particles. 
This phase of the secretory pathway has also been 
reproduced in a cell-free system and has been shown 
to rely on the gene products that are required in 
living cells. The combination of genetic and bio- 
chemical approaches offers the prospect that each 
of the protein molecules involved in the secretory 
pathway may be isolated in pure form and examined 
in simplified reactions. 
Assistant Investigator James M. Cunningham, M.D. 
(Brigham and Women's Hospital) and his colleagues 
have investigated the properties of a transporter of 
cationic amino acids that serves as a receptor for 
leukemogenic retroviruses in mice. They have 
shown that susceptibility to infection and uptake of 
cationic amino acids is reduced in virus-infected 
cells as a consequence of intracellular binding of 
the transporter to newly synthesized envelope. In 
addition, they have identified two related proteins 
that are also transporters of cationic amino acids but 
demonstrate different kinetic properties and tissue 
expression. They are studying the importance of one 
of these transporters in supplying arginine to macro- 
phages that produce nitric oxide, a molecule that 
can mediate communication between cells and en- 
hance the host defense against intracellular patho- 
gens and tumor cells. The production of nitric oxide 
in cells infected by murine leukemia virus is also 
under investigation. 
The five genes encoding subunits of the DNA poly- 
merase III holoenzyme that were previously identi- 
fied by Assistant Investigator Michael E. O'Donnell, 
Ph.D. (Cornell University Medical College) and his 
colleagues have now been sequenced and cloned 
into expression vectors. The subunits were overpro- 
duced and purified in 1 00-mg quantities. Biochemi- 
cal analyses have distinguished individual assays for 
each subunit, and physical studies have elucidated 
many of the principal subunit-subunit contacts in 
the holoenzyme. Sequence comparisons and the lim- 
ited functional information available in the less- 
advanced eukaryotic systems suggest that the multi- 
protein chromosomal polymerase of both yeast and 
humans is quite similar to the Escherichia co/z poly- 
merase III holoenzyme. The large amount of protein 
generated by Dr. O'Donnell's group has made possi- 
ble a collaboration with Dr. John Kuriyan's HHMI 
laboratory at Rockefeller University on the x-ray 
structure of the holoenzyme. The structure of the /3 
clamp was solved, and its ring shape fits nicely with 
the biochemical prediction of a doughnut appear- 
ance. In another project with Epstein-Barr virus, the 
EBNAl origin-binding protein was found to loop out 
the origin DNA, implying that activation of replica- 
tion may be similar to transcription activation, 
which involves loops induced by enhancer binding 
proteins. EBNAl was also found to distort the DNA 
helix at the origin, presumably an early step in pre- 
paring the DNA for the host replication machinery. 
Investigator H. Ronald Kaback, M.D. (University 
of California, Los Angeles) and his colleagues are 
studying the lactose permease of Escherichia coli, 
an extensively characterized membrane protein 
with 1 2 transmembrane a-helical domains. The per- 
mease is a paradigm for a large class of transport 
proteins that utilize energy to drive the accumula- 
tion of sugars, amino acids, and other substances 
against concentration gradients. Although it is par- 
ticularly difficult to obtain high-resolution struc- 
tural information with this class of proteins, recent 
experiments have begun to provide clues regarding 
the arrangement and interaction of the helices in 
lactose permease. Previously the laboratory had con- 
structed a mutant form of the lactose permease in 
which all of the neutral cysteine residues were re- 
placed by other neutral residues, with very little loss 
of activity. New variants of this mutant now have 
been made in which each of the eight charged resi- 
dues (four positive and four negative) that are 
thought to be located within the transmembrane he- 
lices were replaced by a cysteine residue, and they 
are all completely inactive. However, when all pos- 
sible combinations of negatively and positively 
charged amino acid residues are replaced pairwise 
with cysteine, two double mutants exhibit signifi- 
cant activity. Although there are important phenom- 
enological diff^erences between the two sets of dou- 
ble mutants, in both instances a negatively and a 
positively charged residue clearly interact function- 
CELL BIOLOGY AND REGULATION 7 
